Measuring Ad-Evoked Love

ABSTRACT - A 12-item measure for one of the most important human emotions, love, is developed by bringing together the three basic components of love: platonic love, sex, and companionate love (more commonly labeled as warmth). A second-order confirmatory factor analysis is specified using four love-centered perfume ads as stimuli to validate the ad-evoked love scale. Results support the plausibility of the proposed measure in representing two categories of love: romantic love (including platonic love and passionate love) and companionate love. The measure fills a current need by providing an appropriate measure for ad-evoked love.


Ming-Hui Huang (2001) ,"Measuring Ad-Evoked Love", in NA - Advances in Consumer Research Volume 28, eds. Mary C. Gilly and Joan Meyers-Levy, Valdosta, GA : Association for Consumer Research, Pages: 295-300.

Advances in Consumer Research Volume 28, 2001     Pages 295-300


Ming-Hui Huang, National Chung Cheng University


A 12-item measure for one of the most important human emotions, love, is developed by bringing together the three basic components of love: platonic love, sex, and companionate love (more commonly labeled as warmth). A second-order confirmatory factor analysis is specified using four love-centered perfume ads as stimuli to validate the ad-evoked love scale. Results support the plausibility of the proposed measure in representing two categories of love: romantic love (including platonic love and passionate love) and companionate love. The measure fills a current need by providing an appropriate measure for ad-evoked love.


The theme of love is commonly used in advertising, reflecting its importance in consumers’ lives. Lazarus and Lazarus (1994, p. 107) said that when people are asked to name the most important human emotions, love is usually the most important. However, our understanding of the role that love plays in advertising is hindered because various subtypes of love can be evoked by ads, and because current measures lack the specificity to tap into these subtypes of love. We have platonic and passionate love for our partners; however we have companionate love for parents, children, friends, and mankind in general. The soap opera type campaign for Taster’s Choice instant coffee which portrayed an ongoing romance between two neighbors presented the platonic version of love, with the coffee playing a background role (Belch and Belch 1998, p. 273; Shimp 1997, pp. 270-272). A recent ad for the Peugeot 306 automobile, which depicted the wife of a young married-with-children family lovingly looking on as her husband washed the car, and then the couple passionately making love, presents the passionate version of love. The AT&T campaign which showed images of family, love, hope and trust as facilitated by AT&T’s communication technologies (Cleland and Gleason 1997), is likely to produce the companionate version of love (Belch and Belch 1998, p. 269).

These subtypes of love differ in nature. However, the current measurements of ad-evoked love tend to treat the construct of love at too broad a level. For example Aaker, Stayman, and Hagerty’s (1986) warmth monitor used love and tenderness as two major anchor points. To identify feelings elicited by advertising, Aaker, Stayman, and Vezina (1988) clustered together feelings of love, affection, compassion, romance, warmth, tenderness, and warm-heartedness. Edell and Burke (1987) and Burke and Edell (1989) identified warm feelings as one of the three ad feeling dimensions. This dimension included feeling affectionate, calm, concerned, contemplative, emotional, hopeful, kind, moved, peaceful, pensive, sentimental, touched, and warmhearted.

Love in advertising and in consumers’ lives is interwoven with other emotions and can seldom be examined systematically as a distinct emotion. Without isolating love from other emotions, this lack of specificity renders love as a response to a very wide range of stimuli and situations, with consequent difficulty in using the construct for specific strategic purposes. Based on the distinction into two broad categories of love: romantic and companionate love, and by further defining the category of romantic love into platonic and passionate subtypes, this study provides a definition of each subtype and explores their relationship. The purpose of the study reported here is to develop a measure for ad-evoked love, taking its subtypes into account.


Love is a pleasant emotion accompanied by varying degrees of sexual attraction, and is experienced in various relationships. As a higher order construct, love is comprised of a number of variants. The most important and interesting contrast is that between romantic and companionate love (Lazarus and Lazarus 1994, p. 109). Romantic love includes platonic and passionate subtypes, both of which are experienced in romantic relationships with the passionate subtype including strong sexual implications. Companionate love is experienced in various relationships, and excludes the possibility of sexual implications. The following delineates the four love-related components using two differentiators: the degree of sexual implications and the nature of the relationships where the love is embedded. Figure 1 illustrates the relationship of the four related constructs.

Romatic Love: Platonic and Passionate Love

Romantic love is a pleasant emotion that is experienced in romantic relationships. Sex or the possibility of sex is one of the important characteristics of romantic love (Hendrick and Hendrick 1992, p. 24; Shaver and Hazan 1988). However, not all forms of romantic love that are experienced in a romantic relationship are sexual in origin or purpose. The platonic version of romantic love does not involve explicit sexual implications, whereas the passionate version of romantic love does.

Platonic Love. Romantic love can exist without sexual involvement or passion (Lazarus and Lazarus 1994, p. 108). Originally defined by Plato, this subtype of love can be considered as the soul’s dynamic attempt to achieve oneness with the source of its being, whereas sex is but a means of propagating the race and little more (Murstein 1988, p. 22). The campaign for Taster’s Choice instant coffee serves as an example of platonic love, where a romance between the two partners develops by the aid of coffee with only a very limited amount of sexual implication involved.

Sex. Sex is an innate human drive evolved from reproductive requirements (Tompkins 1984, pp. 164-164). Sexual desires are among the strongest motivators of human behavior (Shaver, Hazan, and Bradshaw 1988) and sexual infatuation can result from passionate arousal in the absence of the intimacy and commitment components of love (Sternberg 1988). Ads for Travel Fox sports shoes serve as examples of sexual appeals. The ads showed, from the chest down, two adults engaged in sexual intercourse, completely naked except for the advertised sports shoes and matching socks (Severn, Belch, and Belch 1990).

Passionate Love. Evidence of passionate love involving explicit sexual implications can be shown from the following studies. Branden (1988) stated that people who are happily in love are inclined to experience sexual intimacy as an important vehicle of contact and expression. In developing scales for consumption-related emotions, Richins (1997) measured romantic love using the items sexy, romantic love, and passionate, suggesting sex as a component of passionate love. The ad for the Peugeot 306 serves as an example, with the wife of a young married-with-children family looks fondly at her husband’s car-washing, after which the couple passionately make love.



Companionate Love

Companionate love, commonly known as the warmth construct, is a pleasant emotion that can be experienced in various relationships without involving any sexual implications. MacDonald (1992) defined warmth as a positive emotion that evolved to facilitate cohesive family relationships and paternal investment in children. He considered warmth to be essential in parent-child relationships, and to a lesser extent, in relationships of friendship among peers. Aaker, Stayman, and Hagerty (1986) conceptually defined warmth as a positive, mild, volatile emotion, involving physiological arousal and precipitated by experiencing, directly or vicariously, a love, family, or friendship relationship. Time spent with other people, being reunited with close friends or relatives, being at a party with friends, or going out on a pleasant date are social contexts depicted in ads to generate warmth.

The possibility of sex is prohibited in non-romantic relationships, such as parental or fraternal, as a result of societal rules (Lazarus and Lazarus 1994, p. 114), and thus sexual implications are excluded from companionate love. Lazarus and Lazarus (1994, p. 109) considered that the absence of sexual interest in companionate love distinguishes companionate love from its romantic counterpart. They argue that erotic passion is missing in these variants of companionate love, whereas it holds center stage in romantic love (p.112). Shaver and Hazan (1988) state that parental love differs from romantic love in that sexual attraction and sexual behavior are part of romantic love, while sex is irrelevant to parental love.


As important as love is in both consumers’ lives and advertising, the lack of proper measurement for ad-evoked love makes difficult an assessment of the effectiveness of ads aiming at portraying love. Thus our understanding of the role of love plays in consumer research and advertising is deterred. To fill the gap, this study uses various subtypes of love to develop a reliable and valid measure, taking into account the content characteristics of loving ads.

The Ad-Evoked Love Scale

A pool of eighteen items was collected from a variety of sources to tap into the three basic components of ad-evoked love: platonic love, sex, and companionate love. The subscale of passionate love is achieved by summing the subscales of platonic love and sex.

Sources of the item collection contain measures of emotions in psychology, advertising, and consumption. Items obtained from advertising studies include the warm/tender cluster identified in Aaker, Stayman, and Vezina’s (1988) feelings elicited by advertising, and the warm dimension of Edell and Burke’s (1987) three-dimension ad feeling scales. Items collected from psychological studies include the love cluster in Shaver, Schwartz, Kirson, and O’Connor’s (1987) prototype view for the psychology of emotion, and the pleasure and arousal factors of Russell and Mehrabian’s (1977) three-factor theory of emotions. Items taken from consumption studies are Richins’ (1997) romantic love and love consumption emotions.

The final 18 items that constitute the ad-evoked love scale and which were subjected to the subsequent empirical testing are: sexy, romantic, passionate, intimate, in love, affectionate, erotic, lustful, infatuated, sexually attractive, desired, sentimental, warmhearted, tender, warm, loving, caring, and companionate. Respondens were asked to indicate the intensity of the emotion they felt while exposed to the ad using a five-point scale ranging from "very slightly or not at all" to "extremely."


Four perfume print ads were developed to generate four subtypes of love using two dimensions derived from the previous discussion. The dimensions were the degree of sexual content and the nature of relationships. The four subtypes of love that were manipulated were consistent in format. The platonic loving ad has low sexual content in a romantic relationship, the passionate loving ad depicts high sexual content in a romantic relationship, the companionate loving ad pictures low sexual content in a parental relationship, and the sexual ad portrays high sexual content without a relationship (see Figure 1). The common elements for all ads are a dominant photo depicting the intended content characteristics of each type of love and two bottles of the advertised perfume in the lower part of the ad. For the romantic ads (the platonic and the passionate ads), a pair of wedding rings is in the upper corner of the ads to imply the couple being in a romantic relationship. The platonic ad has a close-up of a tastefully clothed, demure couple smiling. The sexual ad shows a close-up of a nude couple in a sexually suggestive pose. The passionate ad describes a just-married, scantily-dressed couple passionately kissing. The companionate ad depicts a father and son playing chess in a park.

Perfume was selected as the test product since it is a "feeling" product (Erevelles 1998) which often contains both loving and sexual materials. An English perfume brand "After Six," which is only available at UK specialty stores, was used for its low familiarity to the respondents, thus avoiding idiosyncratic reactions to a familiar brand.


Respondents for the study were 215 undergraduate students from a variety of major fields in a California university, who took part in the study to obtain extra course credit. One hundred and eighteen were men and 97 were women, with a mean age of 22.82 (SD = 4.45). Each respondent received a color-printed magazine-like booklet, which contained the four focal ads, mock ads and questions exploring consumption behavior irrelevant to the current study. The order of the ads was counterbalanced and rotated to minimize order effects and carryover biases. The booklets were distributed in classrooms and respondents were instructed to read the booklet as if they were reading a magazine, and to answer the questions independently. The feelings of love generated by these ads were measured right after each ad exposure.


Exploratory Factor Analysis

Principal component exploratory factor analysis with varimax rotation, extracting a three-factor solution, was first used to examine the dimensionality of love generated by the four ads, as measured by the 18-item loving scale. The application of Cattell’s scree test confirmed this three-factor solution because eigenvalues became homogeneous and were smaller than 1.0 after the third factor. The results confirmed the distinct nature of the platonic love, sex and companionate love components. The first companionate love factor loaded the seven items of warmhearted, warm, tender, caring, loving, sentimental, and companionate, accounting for 48.7% of the variance in ad-evoked love. The second sex factor loaded the six items of lustful, erotic, infatuated, sexual attraction, desired, and sexy, accounting for 28.9% of the variance. The third platonic love factor loaded the five items of romantic, in love, affectionate, intimate, and passionate, accounting for 5.6% of the variance.

Second-order Confirmatory Factor Analysis

A second-order confirmatory factor analysis (CFA) with a maximum likelihood fitting function was then specified to confirm the three-component solution obtained from the exploratory factor analysis and to take into account the hierarchical structure of ad-evoked love. This approach has the following two advantages. First, it provides a more rigorous check of the appropriateness of the ad-evoked love scale items to measure their corresponding latent loving constructs than its exploratory counterpart. Second, the second-order latent factor, love, can be specified to examine its effect on the first-order latent factors, platonic love, sex and companionate love, thus examining whether or not those components of love belong to the broad emotional category of love.

An iteration procedure was adopted to select the best scale items. To provide better discriminant validity for each first-order factor, scale items loaded heavily on more than one factor were dropped from the analysis. After this procedure, four items remained for each factor, which all demonstrated satisfactory reliability. Figure 2 represents the model. The single second-order factor is love (x1) that directly influences the three first-order factors: platonic love (h1), sex (h2), and companionate love (h3). These first-order factors have direct effects on their corresponding scale items. The Ly is 12 x 3, with one scale item per construct chosen to scale the latent first-order factors. Similarly, the first element in G scales the second-order factor love (x1) to platonic love (h1), with the remaining second-order factor loadings that are free. This procedure gives the first-order factors the same scale as the observed items, and the second-order factor the same scale as the first-order factors. The Y matrix provides the variance in the first-order factors not explained by the second-order factor.


All first- and second-order factor loadings (Ly and G) are statistically significant and positive. The R2yi’s for the scale items range from .68 to .89, and the R2hi’s are 1.00 (fixed), .39, and .17, respectively. Though these are somewhat good, the chi-square estimate of 789.70 with 52 df is highly significant (p < .000). Moreover the other measures of overall fit suggest that some improvements could be made (e.g., RMSEA = .13). A screen of the modification indices of Ly’s suggests that the first-order factor, platonic love (h1), has a direct effect on the scale item #desired,’ and the first-order factor, sex (h2), has a direct effect on the scale item #intimate.’ The two parameters are thus relaxed and the chi-square estimate is significantly improved to 382.92 with 50 df (p < .000). The RMSEA is at .088, NFI at .96, and CFI at .97, all three overall model fit indices suggesed the appropriateness of the hypothesized model.

The composite reliability estimates for platonic love, sex, and companionate love were .921, .928, and .942, and the estimates of variance extracted were .717, .744, and .802, respectively, both indices suggesting satisfactory reliabilities and variance accounted for by those scale items (Bagozzi and Yi 1988).

Posteriori Manipulation Check

A posteriori manipulation check was conducted to examine whether there was a one-to-one correspondence between the ads’ content and the feelings of love generated by them. The check could not be made in advance due to the undetermined nature of scale items describing the specific type of love. By summing up the four items of each component of love, with passionate love further averaged from the platonic love and sex, the average scores were subjected to analyses of variances with posteriori Tukey-HSD tests. The mean levels of platonic love generated by platonic loving, passionate loving, and sexual ads were homogeneous, but were significantly higher than the companionate loving ad (3.23, 3.70, 3.00 versus 1.53), F = 183.001, p = .000, suggesting the distinctiveness of the companionate love from the romantic love. The sexual ad was rated as being sexier than the passionate, the platonic, and the companionate loving ads (3.38 versus 2.86, 1.89, 1.14), F = 214.081, p = .000. The mean levels of companionate love generated by the companionate, the passionate, and the platonic loving ads were no different (3.13, versus 3.02, 3.23), but were distinct from the sexual ad (2.11), F = 46.644, p = .000. This strengthens the view that the example of companionate love does not contain a sexual component. The passionate loving ad was higher on passionate loving than the platonic and the companionate loving ads (3.28 versus 2.56, 1.34), F = 213.968, p = .000, but it was as passionate as the sexual ad (3.28 versus 3.19), confirming sex as their common component.





This study, reporting results using a second-order CFA analysis, developed a set of loving scales that represent the three major subtypes of ad-evoked love - platonic love, passionate love, and compassionate love - through the three basic components of love - platonic love, sex, and companionate love. The current ad-evoked love scales possess a higher degree of discriminant validity to determine which specific type of love is elicited by ads than previous ad feeling scales.

Measuring Romantic Love

The measurement of romantic love can be achieved through different combinations of the subscales of platonic love and sex. The four core items of romantic, in love, affectionate, and intimate, can each assess platonic love. The fifth item #desired’ is included to provide a complete measure. Summing the subscales of platonic love and sex yields the subscale of passionate love. These subscales are romantic, in love, affectionate, intimate, erotic, lustful, infatuated, and desired. The coefficient alpha for this 8-item set is satisfactory at .909. Compared with Richins’ (1997) romantic love scale, this set of items has the advantage of being able to distinguish the two subtypes of romantic love, depending on whether or not the subscale of sex is included. Thus it provides a more specific measure for love to various loving advertising appeals. For example, the items of #sexy’ and #passionate’ used in Richins’ (1997) romantic love scale might not be appropriate for measuring romantic appeals that do not involve explicit seual implications. A typical example is the Taster’s Choice ad series where a romance is developed without involving explicit sex or passion.

Measuring Companionate Love

To measure companionate love, the four items of sentimental, tender, warm, and caring are recommended. This companionate love scale set has an advantage over the warmth measure used in past advertising studies in that it is neither compounded with the higher order construct of love nor intertwined with the related but distinct construct of platonic love. Compared to the love subscale developed by Richins (1997), which includes items of loving, sentimental, and warm hearted, the current set does not involve the item of loving, an item that can be used to measure all love-related constructs and is thus lacking in discriminant validity. Compared with the warm/tender cluster identified by Aaker, Stayman, and Vezina (1988), this companionate loving scale provides a more specific measure to the construct of warmth by dropping the fringed items, such as sympathetic and empathetic; or the compounding items, such as love. Compared with the warm feelings obtained in Edell and Burke (1987) and Burke and Edell (1989), this scale distinguishes the emotion of companionate love from the arousal dimension of emotion (e.g., Russell and Mehrabian 1977) by dropping items such as calm and peaceful. It is also clarified by dropping cognitive-oriented items, such as contemplative and hopeful, thereby providing a more valid measure for the emotion of companionate love.




Aaker, David, A. and Douglas M. Stayman (1989), "What Mediates the Emotional Responses to Advertising? The Case of Warmth," in Cognitive and Affective Responses to Advertising, eds. P. Cafferata and A. Tybout, Lexington, MA: Lexington Books, 287-303.

Aaker, David and Richard Vezina (1988), "Identifying Feelings Elicited by Advertising," Psychology & Marketing, 5(Spring), 1-16.

Aaker, David and Michael R. Hagerty (1986), "Warmth in Advertising: Measurement, Impact, and Sequence Effects," Journal of Consumer Research, 12(March), 365-381.

Bagozzi, Richard P. and Youjae Yi (1988), "On the Evaluation of Structural Equation Models," Journal of the Academy of Marketing Science, 16(1), 74-94.

Belch, George E. and Michael A. Belch (1998), Advertising and Promotion: An Integrated Marketing Communication, Boston, MA: Irwin/McGraw-Hill, 264-295.

Branden, Nathaniel (1988), "A Vision of Romantic Love," in The Psychology of Love, Robert J. Sternberg and Michael L. Barnes, eds., New Haven: Yale University Press, 218-231.

Burke, Marian Chapman and Julie A. Edell (1989), "The Impact of Feelings on Ad-Based Affect and Cognition," Journal of Marketing Research, 6(February), 69-83.

Cleland, Kim and Mark Gleason (1997), "AT&T Brand Revamp Gets First Support in New Ads," Advertising Age, March 3, 3, 37.

Edell, Julie A. and Marian Chapman Burke (1987), "The Power of Feelings in Understanding Advertising Effects," Journal of Consumer Research, 14(December), 421-433.

Erevelles, Sunil (1998), "The Role of Affect in Marketing," Journal of Business Research, 42, 199-215.

Hendrick, Susan S. and Clyde Hendrick (1992), Romantic Love, Newbury Park, CA: Sage.

Lazarus, Richard S. and Bernice N. Lazarus (1994), Passion and Reason: Making Sense of Our Emotions, New York, NY: Oxford University Press, 86-115.

MacDonald, Kevin (1992), "Warmth as a Developmental Construct: An Evolutionary Analysis," Child Development, 63(August), 753-773.

Richins, Marsha L. (1997), "Measuring Emotions in the Consumption Experience," Journal of Consumer Research, 24(September), 127-146.

Russell, James A. and Albert Mehrabian (1977), "Evidence for a Three-Factor Theory of Emotions," Journal of Research in Personality, 11, 273-294.

Severn, Jessica, George E. Belch and Michael A. Belch (1990), "The Effects of Sexual and Non-sexual Advertising Appeals and Information Level on Cognitive Processing and Communication Effectiveness," Journal of Advertising, 19(1), 14-22.

Shaver, Phillip R. and Cindy Hazan (1988), "A Biased Overview of the Study of Love," Journal of Social and Personal Relationships, 5, 473-501.

Shaver, Phillip R., Cindy Hazan and Donna Bradshaw (1988), "Love as Attachment: The Integration of Three Behavioral Systems," in The Psychology of Love, Robert J. Sternberg and Michael L. Barnes, eds., New Haven: Yale University Press, 68-99.

Shaver, Phillip R., Judith Schwartz, Donald Kirson and Cary O’Connor (1987), "Emotion Knowledge: Further Exploration of a Prototype Approach," Journal of Personality and Social Psychology, 52(6), 1061-1086.

Shimp, Terence A. (1997), Advertising, Promotion, and Supplemental Aspects of Integrated Marketing Communication, Orlando, FL: The Dryden Press.

Sternberg, Robert J. (1988), "Triangulating Love," in The Psychology of Love, Robert J. Sternberg and Michael L. Barnes, eds., New Haven: Yale University Press, 119-138.

Tompkins, Silvan S. (1984), "Affect Theory," in Approaches to Emotion, Klaus R. Scherer and Paul Ekman, eds., Hillsdale, NJ: Lawrence Erlbaum Associates, 163-195.



Ming-Hui Huang, National Chung Cheng University


NA - Advances in Consumer Research Volume 28 | 2001

Share Proceeding

Featured papers

See More


Perceptions of Disability in the Marketplace: Moral Character Inferences and Persuasion

Helen van der Sluis, Arizona State University, USA
Adriana Samper, Arizona State University, USA
Kirk Kristofferson, Ivey Business School

Read More


An Idea Opposed to Another Idea is Always the Same Idea: Reconsidering the Materialistic Aspects of Voluntary Simplicity

Mathieu Alemany Oliver, Toulouse Business School
Justyna Kramarczyk, Adam Mickiewicz University in Poznan

Read More


A Rational Model to Predict Consumers’ Irrational Behavior

Vahid Rahmani, Rowan University

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.